Intellectual property and controversy

Transfer pricing of intellectual property (IP) has always been a difficult technical area but, until recently, guidance on the subject was sparse. The 2010 OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations, Chapter VI, deals with special considerations for the transfer pricing of intangibles in just 39 paragraphs. To put that in context, the same edition takes 42 paragraphs to deal with intragroup services.

Now, under Action 8 of the Base Erosion and Profit Shifting (BEPS) initiative, the OECD has published a revised Chapter VI and material to insert into Chapters I and II, which is intended to be adopted as formal guidance once the work on profit splits and financial services is completed, possibly in early 2017. At 612 paragraphs, plus an annex of examples and the proposed changes to Chapters I and II, there is now a substantial body of guidance on the subject.

Before the revised guidance is adopted into the revised version of the transfer pricing guidelines, it is possible that some countries will adopt the current material into their domestic transfer pricing law. For example, the UK could do so by a simple Order of the Treasury.

Even if not adopted into law, the material is considered, in the main, to be a "better explanation" of the arm's-length principle, and even though it cannot be cited as precedent, the ideas embodied in the guidance may be used immediately. In the authors' experience, several countries, including Australia, Finland, France, the UK, and the US are currently using at least some of the BEPS concepts in audits.

The time frame for completion of work on the BEPS actions was politically inspired and rather short, whilst the other matters also considered under the BEPS project reduced the resources devoted to transfer pricing work. As a result, the final output from Action 8 lacks clarity and therefore may increase controversy rather than reduce it.

This article examines how to interpret the new guidance in accordance with the stated aim of BEPS Actions 8-10: to align transfer pricing outcomes with value creation by proper application of the arm's-length principle, and any controversy that may arise because of the new guidance. It focuses on IP definition, contractual positions as compared to actual activity, and the control of activities and risk. However, before looking at Chapter VI, we must consider the arm's-length principle and the changes proposed for Chapter I of the transfer pricing guidelines.

The arm's-length principle is the almost universal tenet used by tax administrations around the world to guide the allocation of income among affiliated members of a multinational enterprise.

Transfer pricing disputes between countries that are parties to a tax treaty typically rely on Article 9 of the Model Treaty Convention, for resolution on the interpretation of the arm's-length principle articulated in Chapter I of the transfer pricing guidelines. The ability of this "principle" to achieve that purpose depends on its articulation in the transfer pricing guidelines, and the interpretation of the guidelines will be updated substantially as a result of the BEPS process. In this section, we will first distinguish between assessing the arm's-length nature of a transaction and assessing the arm's-length price of that transaction. We will then discuss the two different ways the arm's-length principle can be interpreted insofar as pricing is concerned, and we will comment on its articulation in the expected revision to Chapter I of the transfer pricing guidelines.

A frequent comment at the OECD public consultations on transfer pricing matters and in transfer pricing audits is that "third parties would never enter into this transaction". This statement is typically used to discredit the validity of a transaction that takes place between affiliated members of a multinational enterprise. The implication of the assertion is that the transaction should be recharacterised as one that third parties would actually enter into. Unfortunately, that reasoning is flawed.

Multinationals form because the vertical integration of business assets and the resulting efficiencies, both economic and managerial, from common control provide more value to shareholders than the alternative of fragmented control over the same business assets. As a result, multinational enterprises can and do enter into certain transactions that third parties operating in the open market could not, or would not, enter into. For example, a parent company rich in valuable intellectual property would have a strong incentive to transfer rights to use and possibly further develop all of its intellectual property to an affiliated foreign entity if that transfer of rights increases value to the shareholders. Clearly, this would not necessarily be the case in a third-party context. It is not often that a company licenses all its IP to a competitor. Should tax administrations be allowed to prevent a parent company from transferring various rights to its most valuable IP to affiliated members on the grounds that third parties do not enter into comparable transactions? An affirmative answer would mean that, purely for tax reasons, we are willing to forgo some of the basic economic reasons why multinational companies are socially and economically desirable–they increase the efficiency of the markets and the resulting social welfare.

Statements such as "third parties would never enter into this transaction" are intrinsically subjective and more often than not grounded in an individual's beliefs, as opposed to thorough empirical research. In that case, the arm's-length principle should not be invoked to re-characterise a transaction. Indeed, a principle is a "standard" in that the outcome of the application of the principle must be independent from the identity of the party applying the principle. Perhaps the most blatant example of this is the assertion that a party investing in a development project would not agree to take an equity position in the development project unless it exercised control of the risks associated with the project. If this statement sounds familiar; it should. It is the argument used by the OECD under the BEPS initiative to justify providing low-functionality funding entities with a financial return instead of the upside and downside of the intangible return. The problem with that assertion is that it clearly breaches the arm's-length principle, in that it is easy to find empirical evidence to the contrary. In the movie industry, for example, third-party studios will agree to share all development costs of a movie and share the profit accordingly, with one studio contractually agreeing to relinquish all creative control over the making of the movie to the other studio. So much for exercising control as a prerequisite to share in the upside and downside of the venture.

The revisions to Chapter I of the transfer pricing guidelines may be misinterpreted as empowering tax administrations to test a written contract between affiliated members of a multinational enterprise using their own beliefs and commercial judgment about the rationality of the subject transaction. Specifically, even if one accepts the premise that "control of risk" – at arm's-length – governs who is entitled to an equity return (an erroneous premise, as demonstrated above), the lack of clear guidance, the ambiguity of what controlling risk really means in the context of a multinational enterprise, and most importantly a very clear reference to the arm's length principle – what independent parties actually do – are likely going to result in increasing controversy about the real character of controlled transactions.

Under a behavioural interpretation of the arm's-length principle, the price used between affiliated members of a multinational enterprise in a transaction is arm's-length if it is consistent with the price used by third parties in the open market in a comparable transaction. This interpretation has significant implications on the valuation method used to price transfers of intellectual property. Indeed, to operationalise such interpretation requires that we find comparable transactions in the open market conducted under comparable facts and circumstances. The challenge therefore consists in finding transactional comparables that may not exist in the open market. Because the outcome of third-party transactions occurring in the open market is arguably independent from the identity of those observing such transactions, we have an acceptable standard.

Under US law, the best method rule provides that the most reliable measure of an arm's length result must be used when determining whether a controlled taxpayer's results are consistent with those that would have been realized if uncontrolled taxpayers had engaged in the same transaction under the same circumstances. Although there is no strict priority of methods, recent court cases have affirmed that third-party behavioral evidence of pricing trumps any other interpretation of the arm's length principle.

Under a thought experiment interpretation of the arm's-length principle, economic principles are used to divine how third parties operating in the open market would have priced a transaction that takes place within a multinational enterprise. The economic principles used are typically those addressing the forces of supply and demand in the market, or general cooperative or non-cooperative game theoretic principles, including those of bargaining theory. The thought experiment thus consists in a mental exercise of applying well-accepted principles to analyse a subject transaction. Because the outcome of the application of such principles is arguably independent from the identity of those applying them, we have an acceptable standard. However, from a practical standpoint, this is far from the case. Rigorous application of the arm's-length principle based on a thought experiment requires deep understanding of how competitive markets work, and how cooperating agents and non-cooperating agents reach outcomes. Most transfer pricing practitioners have very little or no training in economics and are either accountants or lawyers by training. Even within the economic profession, most practicing economists have been trained in the behavioural application of the arm's-length principle, not in the thought experiment interpretation.

One example of a thought experiment-based application of the arm's-length principle is the use of corporate finance valuation techniques to price transfers of intellectual property. That methodology consists of projecting a probability-weighted forecast (first thought experiment), selecting various valuation parameters that capture opportunity costs, for example (second thought experiment), and adding some version of the risk-adjusted projections into a present value (third thought experiment) under the overall premise that this is the way a third party would have assessed the development activity (speculative assertion). The revisions to Chapter VI of the transfer pricing guidelines introduce such valuation methods for the first time and provide extensive guidance on how to apply them.

The changes to Chapters I and VI of the transfer pricing guidelines substantially tilt the interpretation of the arm's-length principle from a behavioural interpretation to a thought experiment-based interpretation. Thought experiment-based interpretations of the arm's-length principle are intrinsically more subjective and arbitrary than behavioural interpretations. Having said that, because multinational enterprises often engage in transactions that uncontrolled participants would not enter into, it may well be the case that a thought experiment is the only choice to interpret and operationalise the arm's-length principle. However, in cases in which both interpretations are possible, it is likely that under the new guidance of Chapter I of the transfer pricing guidelines a thought experiment interpretation will be used by the tax administrations and thus result in increased controversy.

The revisions to Chapter I of the transfer pricing guidelines and to Chapter VI dealing with transfers of intangibles are voluminous, complicated, and susceptible to accommodating various interpretations that can lead to very different allocations of income. In connection with the transfer of intellectual property, the most telling example of the previous assertion is the guidance providing that a low-functionality entity that controls only the funding risk is entitled to no more than a risk-adjusted return. The guidance, however, is silent as to what a risk-adjusted return consists of, and no examples are provided to help understand how to determine what a risk-adjusted return is. Ask an economist what a risk-adjusted return is and you will hear that it is a risk-free rate of return augmented by a premium reflecting at least default risk and possibly other premiums reflecting other risks, such as the diversifiable or non-diversifiable risks of the investments. The term "risk-adjusted" is so generic and uninformative that it clearly reflects the lack of consensus among OECD member countries as to exactly how much income should be provided to a low-functionality entity controlling the funding and no other risks. When the lack of consensus as to what the outcome ought to be translates into guidance that is so vague that it can accommodate very different views of what that outcome should be, dispute and controversy are the likely outcomes.

If the aforementioned sounds problematic, consider that none of the guidance provided in the revisions to the transfer pricing guidelines (Chapters I and VI) deal with ex-post allocations of income. The revisions address the ex-ante allocation of income but are silent as to who, in the multinational enterprise, is entitled to the realised upside and downside of the business. Certainly the introduction into Chapter VI of measures that match the US "commensurate with income" provisions addresses to some degree the ex-post allocation of income, but it does so to resolve ex-ante informational asymmetries between taxpayers and tax administrations. As such, it is not really dealing with unexpected realisation of upsides and downsides with regard to an ex-ante valuation that was respected by tax administrations.

Although tax administrations are likely to contest allocating unexpected upsides to low-functionality entities that just provide and control the funding, typical multinational enterprises include a number of participants in the so-called DEMPE functions (development, enhancement, maintenance, protection, exploitation) that arguably would claim the unexpected upsides. We discuss in the valuation methodologies section the increased pressure to use a global profit split to allocate unexpected upsides. It is not clear that tax administrations will apply the same pressure when it comes to splitting unexpected losses but, aside from the valuation issue, there is the more fundamental issue that taxpayers face a significant amount of uncertainty as to what position various tax administrations will take in claiming a right to tax unexpected upsides of intellectual property development and exploitation. The purpose of the transfer pricing guidelines is to provide a framework that alleviates that uncertainty and provides reasonable certainty as to the outcome of transactions, including whether or not they will be respected.

As noted above, the OECD's final report on Actions 8-10 of the BEPS Action Plan indicates that it is mainly a better explanation (or a clarification) of the arm's-length principle and, as such, we should expect taxpayers and tax administrations to start applying the guidance immediately. This does not necessarily mean that all revisions to the transfer pricing guidelines will necessarily be effective immediately. Indeed, we now understand that there will be specific start dates for the guidance on hard-to-value intangibles, low-value-adding services and, we assume, cost contribution arrangements. The choice of dates will be discussed by the OECD's Working Party 6 as part of its work on implementation. This work is expected to take place primarily in 2016.

The revised draft Chapter VI begins by endorsing the arm's-length principle and the guidance in Chapters I-III (para 6.1-3). Therefore, everything that follows in Chapter VI must be read in the context of the arm's-length principle – what actually happens between unrelated parties.

Section A of the revised Chapter VI defines an intangible in paragraph 6.6 as "…not a physical asset or a financial asset, which is capable of being owned or controlled for use in commercial activities, and whose use or transfer would be compensated had it occurred in a transaction between independent parties in comparable circumstances".

Anything falling short of that definition may be taken into account in a comparability analysis, but it is not an "intangible" for purposes of the guidance. In commercial terms, anything meeting these conditions is "intellectual property", but that commercial term is not used. That is unfortunate, as the guidance could be clearer – and more concise – if it did.

The inclusion of the phrase "owned and controlled" as part of the definition of an intangible leads us to consider the legal nature of IP. Because intangibles have no physical form, ownership and control are achieved by the creation of a monopoly over use, which only the state can do, in the form of intellectual property law and/or contract law. Unhelpfully, paragraph 6.8 notes that "legal, contractual, or other forms of protection" may affect the value of an item, but claims that the existence of such protection is not necessary for the item to be characterised as an intangible. However, without "protection" the item cannot be "controlled" and therefore it cannot be an intangible under paragraph 6.6. This confusion is likely to lead to controversy in identifying intangibles.

The potential for controversy may be significantly reduced by identifying intangibles "with specificity", as instructed in paragraph 6.12. To show that the [assumed] intangible is "owned and controlled" (paragraph 6.6) it is necessary to show that a monopoly exists. The converse is also true: If we show that the [assumed] intangible is not "owned and controlled", then we have shown that it is not an intangible for purposes of the guidance. By diligent analysis to produce work of evidential quality, the scope for controversy may be greatly reduced. For example, suppose that A, a US business, has a patent over a process that it licenses to overseas group member B. The patent expires, but B has used that process in developing a new product or service with such novelty that a new, valuable patent is granted to B. It is sometimes argued that the original work and expired patent must give rise to "platform IP" from which the new idea was developed, and so a fee must be paid by B to A even though the patent has expired. As we are applying the arm's-length principle, would unrelated parties pay a fee either as a general matter or because A would include in its original license to B that it must continue to pay even after expiry of the patent? The answer is simply no, no fee would be due. It is a fundamental tenet of US patent law that there is no fee due once a patent expires, and that to contractually provide otherwise is contrary to the intention of IP law. Moreover, such a clause would be unenforceable (see Kimble v Marvel Entertainment LLC (576 U.S. __, 135 S. Ct. 2401 (2015). Hence, we can prove evidentially that there is no concept of "platform IP" and we can avoid controversy over its existence.

Similarly, items that are not commonly understood by tax specialists to be "owned and controlled" may be identified as intangibles by diligent analysis using the relevant law and commercial practice of the pertinent territory to produce work of evidential quality. For example, a UK waste management license, an exemption from a general prohibition against dumping waste, is an intangible (see National Provincial Bank v Ainsworth [1965] AC 1175).

The OECD's new guidance on valuation methods will likely have the effect of increasing the use of the profit split method. The guidance is clear that any of the traditional transfer pricing methods might be useful, or as would be other valuation methods in certain circumstances. However, the guidance dismisses one-sided methods, including the resale price method and the transactional net margin method, as generally not reliable for valuing intangibles. Similarly, the guidance states that cost-based methods are generally discouraged, and points out the difficulty of applying a comparable uncontrolled price (CUP) in practice, leaving only the profit split method available from the traditional transfer pricing methods. In addition, the guidance could be read as favouring income-based methods in the valuation of an intangible asset (as opposed to the license of that intangible).

Comparability issues are likely to arise when considering CUPs. All intangibles that are "owned or controlled" have the commercial characteristic of intangible property, and the third-party comparable licenses or sales that can be found are only of intangible property. To achieve intangible property status, the item must be "unique"; thus, in many cases the item will fail the test of comparability, other than when internally generated comparables are involved. This can be illustrated by commercial law cases. For example, a UK court in 1975 rejected a "comparables search" of IP licenses in an infringement case because comparability had not been shown (see General Tyre & Rubber v Firestone Tyre & Rubber [1975] 1 W.L.R). Sometimes, even though the intangible is unique, it performs a comparable economic function. If there are several options to achieve the same end at the same cost and each can be licensed, then these alternatives provide a comparable uncontrolled price. However, in many cases, no CUP evidence will be available.

In applying the arm's-length principle, it will sometimes be acceptable to use cost-based methods, if that is what happens in unrelated-party transactions. For example, early-stage development intangibles – for which significant risk remains in the development process – are often traded based on the costs incurred to date. When that can be shown to happen between unrelated parties, then the same method must be acceptable for related-party transactions under the arm's-length principle. However, taxpayers seeking to use a cost-based method should ensure their evidence is on point, because tax administrations may point to the guidance which, read incorrectly, may appear to amount to a general prohibition against the use of cost methods.

The guidance in its current state provides practically no help on how to perform a profit split calculation, but further work on profit splits will be completed during 2016, leading to additional guidance in 2017. However, courts in the US and many other countries have been determining what constitutes a "fair license" for intangibles for many years. That body of case law provides evidence on whether a court judges a method to be acceptable or not. For example, in the 2011 case Uniloc LLC v Microsoft Corporation, the US Federal Court of Appeals found that the "rule of thumb" (a basis for a profit split) was not evidential and refused to accept an estimation of a fair royalty based thereon. Courts in the US and other countries have endorsed the use of profit split methods when more evidential approaches are used. Specifically, the full profit split method – as opposed to the residual profit split method – has found favour in many countries. This body of jurisprudence provides valuable evidence of what methods would be acceptable between unrelated parties, which can form a sound basis for transfer pricing valuation.

The stated aim of Action 8 of the BEPS project was to align transfer pricing outcomes with value creation by proper application of the arm's-length principle. The combination of a short time frame and high workload (the OECD's BEPS Action Plan included 15 actions) has resulted in guidance that lacks clarity. Unless transfer pricing professionals keep the arm's-length principle firmly in mind, it is possible to misconstrue the guidance as supporting non-arm's-length outcomes. A lack of understanding of what constitutes an intangible, or a failure to adhere to the arm's-length principle is likely to lead to non-arm's-length outcomes in practice. This will increase the risk of controversy and, subsequently, the number of claims to enter the mutual agreement procedure of a bilateral tax treaty or similar provision of an applicable multilateral treaty, such as the EU Arbitration Convention.

The easiest way to avoid that risk is through the advance pricing agreement mechanism, whenever possible. Although entering into an APA is unlikely to lead to higher costs for the taxpayer, compared to maintaining documentation and undergoing a thorough tax audit, more tax authority resources are required. As yet there is little evidence outside the US that tax administrations are addressing this staffing need.

Whether taxpayers take the APA option or choose instead to document their intangibles' transfer pricing and await a tax audit, it is clear that all work supporting the transfer pricing position should be "evidential" in nature. This is by far the most effective way to control the potential for controversy, and indeed to reduce the cost of an APA. In any event, if agreement is not reached with the tax auditor the matter proceeds to appeal and rules in most countries define the standard of material that can be introduced in court as "expert testimony". In the US this raises the possibility to make a "Daubert motion" against material that is not of "evidential quality" for it to be excluded from consideration.

John Henshall Deloitte 2 New Street Square London EC4A 3BZ Tel: +44 207 303 2218 jhenshall@deloitte.co.uk www.deloitte.com John Henshall is a partner in the UK Transfer Pricing group and global co-lead of the business model optimisation (BMO) services. He specialises in the transfer pricing of business reorganisations and intellectual property, which are significant value drivers for defence and aerospace multinationals. With more than 30 years of tax experience, John has a successful history of achieving strong compliance positions, strong documentation and, when appropriate, both advance pricing agreements and successful tax controversy settlements. Training initially with the UK tax authority, John became a Deloitte UK partner in 2001. John represents Deloitte UK to HMRC at the national level on transfer pricing technical matters, and he represents Deloitte globally at OECD's Working Party 1, dealing with permanent establishment issues, and Working Party 6, dealing with transfer pricing of intangibles. John is consulted by overseas governments on the modernisation of their approach to international taxation and transfer pricing, including base erosion and profit shifting (BEPS) issues. He lectures on supply-chain reorganizations and the transfer pricing of intangibles both to industry and to tax authorities. John is regularly published, including in journals such as Practical US/International Tax Strategies, Tax Planning International: Transfer Pricing, Bloomberg BNA's Transfer Pricing Report, and The Tax Journal. John is the author of Global Transfer Pricing: Principles and Practice (2nd Edition) [Bloomsbury Professional, 2013] and a contributing author on transfer pricing to Ray, Partnership Taxation [Lexis Nexis, Tolley].

Philippe Penelle Principal, Washington National Tax Deloitte 555 12th Street NW Washington DC Tel: +1 213 399 5642 Fax: +1 213 694 5480 ppenelle@deloitte.com Philippe Penelle is a principal with the Washington National Tax office of Deloitte Tax, specialising in designing, valuing, and defending transactions that involve the transfer of intellectual property rights. Philippe brings more than 17 years of professional transfer pricing experience assisting multinational clients to set up, maintain, document, and defend transfers of intellectual property rights through cost sharing arrangements, contributions to international partnerships under Internal Revenue Code §704, contributions to corporations under IRC §367, and licensing arrangements involving specific allocations of fixed-cost funding commitments. His practice includes advising and representing clients in high-profile cost sharing controversy cases involving large amounts of potential adjustments. In addition to his extensive client work in the life sciences, technology, video game, entertainment and media, fashion, and retail industries, Philippe has published a number of articles developing valuation methodologies relevant to various intellectual property structures consistent with the OECD transfer pricing guidelines, the Internal Revenue Code, and the Treasury regulations promulgated thereunder. These articles have been published by Bloomberg BNA, International Tax Review and Global Tax Weekly. Philippe serves as co-chair of the US Council for International Business (USCIB) Transfer Pricing Subcommittee. He is actively involved in the BEPS conversations with the OECD, the US Treasury Department, and the international business community. His involvement has included drafting comments submitted on behalf of Deloitte Tax to the OECD, providing input to the USCIB and the Business and Industry Advisory Committee (BIAC) to the OECD, as well as being an invited speaker at OECD public consultations. Philippe is a frequent guest speaker at international conferences including the OECD's International Tax Conference held annually in Washington DC and is regularly cited and quoted in the tax press. Philippe is recognised by Euromoney/Legal Media Group as one of the world's leading transfer pricing advisers. Philippe earned his PhD in economics from the Department of Economics at the University of Chicago, after earning an MA in Econometrics (Summa Cum Laude) and a License en Sciences Economiques Summa Cum Laude, both from the Université Libre de Bruxelles (Belgium).